Chip Policy: The UK Vs. The US Vs. EU Vs. India

As we’ve observed over our recent blogs, the semiconductor industry is reckoning internally with technological questions on densification, scale, overcoming the memory wall, making data movement more efficient and many other such challenges.

But the power and impact of semiconductors has much higher visibility amongst global powers. Seeing how chips now feature in every diplomatic discussion, governments around the world are establishing or revamping their policies to spur semiconductor innovation or control.

In late April, the UK announced its plans to launch a new AI Hardware Plan at June’s London Tech Week. But Technology Secretary Liz Kendall also outlined a smaller-scale, targeted approach to chip policy—a deliberate choice to differentiate from the US, EU, India, and others.

The UK’s Bet
Kendall explained the Government’s reasoning with hard numbers: the global AI chips market is set to reach $1 trillion by the early 2030s. Capturing just 5% of that market would mean $50 billion in revenue for the UK. That’s plenty of incentive on its own.

In its goal to capture that 5% of market share, the UK is taking a similarly slimmed-down approach to spending. The Sovereign AI initiative, built to invest in UK companies, has a first fund of £500 million ($671.6 million USD). While that’s not pocket change by any metric, it’s a fraction of the capital other polities have targeted.

The original EU Chips Act aimed to capture €43 billion ($50 billion USD) of investment. Since going into effect in 2023, more than €80 billion ($93 billion USD) in capital investments has flowed onto the continent. The US’ CHIPS Act made $53 billion USD available, plus tax credits. India’s Semiconductor Mission 2.0 has set aside Rs 1.2 lakh crore (roughly $12.5 billion USD).

However, looking solely at the numbers focuses on the wrong thing.

Given that fabs can now cost more than $20 billion apiece and several years to bring online, rebuilding a full-stack domestic semiconductor industry (like the US is attempting to do) just isn’t a smart use of capital for the UK. Instead, the UK’s Council for Science and Technology has opted for a realistic approach that plays to the UK’s strengths and reinforces them.

With the legacy of companies like Arm and world-class research universities, Kendall argues that the UK is “better placed than almost any other country to run the race to develop new, quicker, smarter and more energy efficient hardware which is critical for the future of AI and our future prosperity.” And they’re putting their funds towards the kinds of companies set to do that.

The US’ All-In Approach
 The US’ goals, in contrast, are far larger. It’s trying to rebuild manufacturing capacity, secure supply chains for defensive-related purposes, and reduce dependence on Taiwan for advanced production given its perpetually precarious geopolitical situation.

However, what the US is still missing is finishing capacity. TSMC’s new Fab 21 in Phoenix can churn out chips—that still go to Asia for testing, singulation, and packaging. CoWoS packaging, which AI accelerators depend on, is still almost entirely done in Taiwan. The US’ new fabs and growing ecosystem are impressive, yes, but they don’t cover the full manufacturing stack quite yet.

The EU’s Bid for Irreplaceability
As previously noted, the EU’s first Chips Act goal of bringing in investment has succeeded. But now, it appears that their goal is neither sovereignty nor specificity; it’s being irreplaceable. The European Parliament makes that clear:

“The proposal for an EU chips act 2.0, scheduled for publication on 27 May 2026 as part of the Tech Sovereignty Package, will aim to strengthen Europe’s resilience and technological sovereignty in semiconductors. It is expected to address the EU’s lack of manufacturing capacity for advanced semiconductor nodes and for other chip markets where the EU holds a competitive advantage, improve the monitoring of semiconductor markets, and simplify the regulatory framework.”

In short, the EU is looking for a balanced approach. As GlobalFoundries’ general manager and senior VP for Europe, Manfred Horstmann, pointed out to EE Times, European firms do have some critical areas of strength: automotive chips, power semiconductors, payments chips, and specialty processing. Chasing hyperscalers that require hyper-capital doesn’t strategically make sense.

Or, as Horstmann put: “Don’t invest in stuff where we can only lose. Invest in stuff where we can win.”

India’s Globally-Focused Goal
The Indian government launched India Semiconductor Mission 2.0 earlier in 2026, building on the success of ISM 1.0 with more capital and bigger goals. ISM 2.0 explicitly states its goal as: “Deepen domestic semiconductor capabilities at a time when chips underpin every critical digital and industrial system.”

In some ways, this mirrors the US’ approach. But while the US is trying to restore manufacturing, India is building it up. ISM 2.0’s goals include achieving 3nm and 2nm node capability by 2035, expanding the fabless startup base to 50 companies, and building a talent pipeline through partnerships with institutions including Lam Research, targeting 60,000 trained professionals over the next decade.

However, prioritizing “domestic semiconductor capabilities” doesn’t preclude India from also strengthening its international place in the industry. Prime Minister Narendra Modi explicitly named semiconductors as a key facet of India and the EU’s partnership in light of a new free trade agreement and planned investment agreement.

Dutch foreign minister Tom Berendsen put it even more explicitly: “India brings enormous engineering talent, innovation capacity, and growing manufacturing ambitions. That is a very powerful combination [with the Netherlands’ advanced ecosystem.]”

While India is clearly mindful of the need for domestic capabilities as geopolitical pressures continue to mount, it’s also clearly working under the belief that the global ecosystem is undergoing a redistribution, not a contraction, and wants to secure its place in that new ecosystem.

So Which Approach is Best?
No economy runs a complete semiconductor stack, not even the US. Consequently, every policy we’ve discussed is trying to do the same 3 things: Define what they intend to control, where they’re comfortable being interdependent, and how much capital they’re willing to invest for their positioning.

Even the UK’s AI Hardware Plan aims to expand on the strengths it already has, not necessarily trying to completely catch up in all areas.

Every decision made by these governments directly impacts companies operating in this landscape, including us here at Baya. Earlier this year, we both opened our first official UK office and expanded our India offices. We’re not the only ones making sure we have footholds in these key markets, either.

No semiconductor industry company is too big to ignore these policy shifts. Not even the biggest of them all.